The Agent Skills Directory

[COMMAND_EXECUTION]: The script scripts/with_server.py utilizes subprocess.Popen with the shell=True parameter to run commands provided via the --server command-line argument. This allows for the execution of arbitrary shell strings, which is a significant security risk if the input strings are derived from untrusted project metadata or external configuration files.
[PROMPT_INJECTION]: The SKILL.md file contains instructions that discourage the agent from reviewing the source code of the included scripts ('DO NOT read the source until you try running the script first'). This behavior-influencing prompt acts as a safety bypass, as it pressures the agent to execute code as a 'black-box' without verifying its security implications or identifying the risky shell execution patterns within.

webapp-testing